Linked Open Data and Its Potential for CREDO Martin Homola Faculty - - PowerPoint PPT Presentation

▶
linked open data and its potential for credo
SMART_READER_LITE
LIVE PREVIEW

Linked Open Data and Its Potential for CREDO Martin Homola Faculty - - PowerPoint PPT Presentation

Mar 24, 2023 196 likes •662 views

Linked Open Data and Its Potential for CREDO Martin Homola Faculty of Mathematics, Physics and Informatics Comenius University in Bratislava CREDO Week 2018 Martin Homola Linked Open Data and Its Potential for CREDO Knowledge Representation

slide-1
SLIDE 1

Linked Open Data and Its Potential for CREDO

Martin Homola

Faculty of Mathematics, Physics and Informatics Comenius University in Bratislava

CREDO Week 2018

Martin Homola Linked Open Data and Its Potential for CREDO

slide-2
SLIDE 2

Knowledge Representation

Martin Homola Linked Open Data and Its Potential for CREDO

slide-3
SLIDE 3

Knowledge Representation

Knowledge representation (KR) studies how knowledge is captured, represented, and reasoned with in computer systems in order to solve complex tasks.

Martin Homola Linked Open Data and Its Potential for CREDO

slide-4
SLIDE 4

Artificial Agent

Martin Homola Linked Open Data and Its Potential for CREDO

slide-5
SLIDE 5

Artificial Agent in Environment

Martin Homola Linked Open Data and Its Potential for CREDO

slide-6
SLIDE 6

Artificial Agent in Environment

human furniture kitten furniture floor

Martin Homola Linked Open Data and Its Potential for CREDO

slide-7
SLIDE 7

Artificial Agent in Environment

human furniture kitten furniture floor dirt_nearby → action(clean) battery_low → action(go_to_dock) . . .

Martin Homola Linked Open Data and Its Potential for CREDO

slide-8
SLIDE 8

Linked Open Data

Martin Homola Linked Open Data and Its Potential for CREDO

slide-9
SLIDE 9

Can We Make Web Machine Understandable?

Martin Homola Linked Open Data and Its Potential for CREDO

slide-10
SLIDE 10

What Can We Read Out?

What useful information can we read in this page?

Martin Homola is an assistant professor. Martin Homola has e-mail homola@fmph.uniba.sk. Martin Homola teaches Computational Logic.

Martin Homola Linked Open Data and Its Potential for CREDO

slide-11
SLIDE 11

What Can We Read Out?

What useful information can we read in this page?

Martin Homola is an assistant professor. Martin Homola has e-mail homola@fmph.uniba.sk. Martin Homola teaches Computational Logic.

All simple property assertions of the form:

subject predicate object .

Martin Homola Linked Open Data and Its Potential for CREDO

slide-12
SLIDE 12

RDF: Resource Description Framework

The SW language of triples:

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dai: http://dai.fmph.uniba.sk/RDF# . dai:homola rdf:type dai:AssistantProfessor . dai:homola dai:name "Martin Homola" . dai:homola dai:email "homola@fmph.uniba.sk" . dai:complog rdf:type dai:Course . dai:complog dai:title "Computational Logic" . dai:complog dai:instructor dai:homola .

Martin Homola Linked Open Data and Its Potential for CREDO

slide-13
SLIDE 13

RDF: Resource Description Framework (cont.)

Abbreviated syntax:

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dai: http://dai.fmph.uniba.sk/RDF# . dai:homola rdf:type dai:AssistantProfessor ; dai:name "Martin Homola" ; dai:email "homola@fmph.uniba.sk" . dai:complog rdf:type dai:Course ; dai:title "Computational Logic" ; dai:instructor dai:homola .

Martin Homola Linked Open Data and Its Potential for CREDO

slide-14
SLIDE 14

Ontologies

But where is the semantics in the Semantic Web?

Martin Homola Linked Open Data and Its Potential for CREDO

slide-15
SLIDE 15

Ontologies

But where is the semantics in the Semantic Web? It is in ontologies.

Martin Homola Linked Open Data and Its Potential for CREDO

slide-16
SLIDE 16

Ontologies

But where is the semantics in the Semantic Web? It is in ontologies.

◮ Everyone can create properties and classes

. . . with defined meaning

◮ Everyone can read the definitions and understand the data ◮ Everyone can reuse already defined properties and classes

Martin Homola Linked Open Data and Its Potential for CREDO

slide-17
SLIDE 17

Ontologies

But where is the semantics in the Semantic Web? It is in ontologies.

◮ Everyone can create properties and classes

. . . with defined meaning

◮ Everyone can read the definitions and understand the data ◮ Everyone can reuse already defined properties and classes

SW ontology languages: RDF Schema & OWL

Martin Homola Linked Open Data and Its Potential for CREDO

slide-18
SLIDE 18

RDF Schema

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix rdfs: http://www.w3.org/2000/01/rdf-schema# . @prefix dai: http://dai.fmph.uniba.sk/RDF# . @prefix schema: http://schema.org/. dai:Professor rdf:type rdfs:Class ; dai:AssistantProfessor rdf:type rdfs:Class ; rdfs:subclassOf dai:Professor . dai:Course rdf:type rdfs:Class; dai:name rdf:type rdf:Property ; rdfs:domain dai:Professor ; rdfs:range rdfs:Literal . dai:instructor rdf:type rdf:Property ; rdfs:domain dai:Course ; rdfs:range dai:Professor .

Martin Homola Linked Open Data and Its Potential for CREDO

slide-19
SLIDE 19

Semantics: Take Two

RDF is a logic with model-theoretic semantics This enables to characterize:

◮ Consistency of vocabularies and data-sets ◮ Logical consequence ◮ Entailment and derivation ◮ Query answering

Enables algorithmization Polynomial complexity

Martin Homola Linked Open Data and Its Potential for CREDO

slide-20
SLIDE 20

SPARQL Query Language

Query language designed to retrieve and manipulate store triple data

PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> PREFIX dai: <http://dai.fmph.uniba.sk/RDF#> SELECT ?name ?email WHERE { ?person rdf:type dai:Professor ; dai:name ?name ; dai:email ?email . ?course dai:instructor ?person . } GROUP BY ?person HAVING (count(?course) > 3)

Martin Homola Linked Open Data and Its Potential for CREDO

slide-21
SLIDE 21

Link Open Data

Early 2000s:

◮ SW brought a number of useful technologies: RDF/S, OWL,

SPARQL, RIF/SWRL, WSDL, SSWAP , . . .

◮ Implemented RDF triple stores and tools ◮ Few early vocabularies appear

Late 2000s Linked Open Data:

◮ Why not actually start using it?

Martin Homola Linked Open Data and Its Potential for CREDO

slide-22
SLIDE 22

LOD Cloud

Martin Homola Linked Open Data and Its Potential for CREDO

slide-23
SLIDE 23

LOD Cloud

Martin Homola Linked Open Data and Its Potential for CREDO

slide-24
SLIDE 24

LOD Cloud

Martin Homola Linked Open Data and Its Potential for CREDO

slide-25
SLIDE 25

LOD Cloud

Martin Homola Linked Open Data and Its Potential for CREDO

slide-26
SLIDE 26

LOD Cloud

Martin Homola Linked Open Data and Its Potential for CREDO

slide-27
SLIDE 27

Example: DBPedia

Free open RDF database extracted from Wikipedia. Accessible via human readable browser, RDF files and SPARQL interface.

Martin Homola Linked Open Data and Its Potential for CREDO

slide-28
SLIDE 28

Example: DBPedia

Free open RDF database extracted from Wikipedia. Accessible via human readable browser, RDF files and SPARQL interface. DBpedia contains 9.5b triples about 6m entities:

◮ 1.5m people ◮ 810k places ◮ 275k organizations ◮ 410k works ◮ 301k species

Martin Homola Linked Open Data and Its Potential for CREDO

slide-29
SLIDE 29

What Can We Find in DBPedia

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dbr: http://dbpedia.org/resource/ . @prefix dbo: http://dbpedia.org/ontology/ . dbr:Marie_Curie rdf:type dbo:Scientist ; dbo:field dbr:Chemistry ; dbo:field dbr:Physics ;

Martin Homola Linked Open Data and Its Potential for CREDO

slide-30
SLIDE 30

What Can We Find in DBPedia

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dbr: http://dbpedia.org/resource/ . @prefix dbo: http://dbpedia.org/ontology/ . dbr:Marie_Curie rdf:type dbo:Scientist ; dbo:field dbr:Chemistry ; dbo:field dbr:Physics ; dbo:birthPlace dbr:Warsaw ; dbo:birthDate "1867-11-07" ;

Martin Homola Linked Open Data and Its Potential for CREDO

slide-31
SLIDE 31

What Can We Find in DBPedia

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dbr: http://dbpedia.org/resource/ . @prefix dbo: http://dbpedia.org/ontology/ . dbr:Marie_Curie rdf:type dbo:Scientist ; dbo:field dbr:Chemistry ; dbo:field dbr:Physics ; dbo:birthPlace dbr:Warsaw ; dbo:birthDate "1867-11-07" ; dbo:birthName "Maria Salomea Skłodowska" ;

Martin Homola Linked Open Data and Its Potential for CREDO

slide-32
SLIDE 32

What Can We Find in DBPedia

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dbr: http://dbpedia.org/resource/ . @prefix dbo: http://dbpedia.org/ontology/ . dbr:Marie_Curie rdf:type dbo:Scientist ; dbo:field dbr:Chemistry ; dbo:field dbr:Physics ; dbo:birthPlace dbr:Warsaw ; dbo:birthDate "1867-11-07" ; dbo:birthName "Maria Salomea Skłodowska" ; dbo:deathDate "1934-07-04" ; dbo:deathPlace dbr:Sancellemoz ; dbo:deathPlace dbr:Passy,_Haute-Savoie ;

Martin Homola Linked Open Data and Its Potential for CREDO

slide-33
SLIDE 33

What Can We Find in DBPedia

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dbr: http://dbpedia.org/resource/ . @prefix dbo: http://dbpedia.org/ontology/ . dbr:Marie_Curie rdf:type dbo:Scientist ; dbo:field dbr:Chemistry ; dbo:field dbr:Physics ; dbo:birthPlace dbr:Warsaw ; dbo:birthDate "1867-11-07" ; dbo:birthName "Maria Salomea Skłodowska" ; dbo:deathDate "1934-07-04" ; dbo:deathPlace dbr:Sancellemoz ; dbo:deathPlace dbr:Passy,_Haute-Savoie ; dbo:deathCause dbr:Aplastic_anemia ;

Martin Homola Linked Open Data and Its Potential for CREDO

slide-34
SLIDE 34

What Can We Find in DBPedia

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dbr: http://dbpedia.org/resource/ . @prefix dbo: http://dbpedia.org/ontology/ . dbr:Marie_Curie rdf:type dbo:Scientist ; dbo:field dbr:Chemistry ; dbo:field dbr:Physics ; dbo:birthPlace dbr:Warsaw ; dbo:birthDate "1867-11-07" ; dbo:birthName "Maria Salomea Skłodowska" ; dbo:deathDate "1934-07-04" ; dbo:deathPlace dbr:Sancellemoz ; dbo:deathPlace dbr:Passy,_Haute-Savoie ; dbo:deathCause dbr:Aplastic_anemia ; dbo:doctoralAdvisor dbr:Gabriel_Lippmann ;

Martin Homola Linked Open Data and Its Potential for CREDO

slide-35
SLIDE 35

What Can We Find in DBPedia

@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# . @prefix dbr: http://dbpedia.org/resource/ . @prefix dbo: http://dbpedia.org/ontology/ . dbr:Marie_Curie rdf:type dbo:Scientist ; dbo:field dbr:Chemistry ; dbo:field dbr:Physics ; dbo:birthPlace dbr:Warsaw ; dbo:birthDate "1867-11-07" ; dbo:birthName "Maria Salomea Skłodowska" ; dbo:deathDate "1934-07-04" ; dbo:deathPlace dbr:Sancellemoz ; dbo:deathPlace dbr:Passy,_Haute-Savoie ; dbo:deathCause dbr:Aplastic_anemia ; dbo:doctoralAdvisor dbr:Gabriel_Lippmann ; dbo:doctoralStudent dbr:Marguerite_Perey ; dbo:doctoralStudent dbr:Óscar_Moreno ; ...

(see live) Martin Homola Linked Open Data and Its Potential for CREDO

slide-36
SLIDE 36

Scientific Data in LOD

How can science benefit from LOD?

Martin Homola Linked Open Data and Its Potential for CREDO

slide-37
SLIDE 37

Scientific Data in LOD

How can science benefit from LOD?

◮ Scientific datasets are already published in LOD ◮ Specific scientific domain ontologies

Martin Homola Linked Open Data and Its Potential for CREDO

slide-38
SLIDE 38

Scientific Data in LOD

How can science benefit from LOD?

◮ Scientific datasets are already published in LOD ◮ Specific scientific domain ontologies

Especially in :

◮ Medicine ◮ Biology ◮ Linguistics

(see example) Martin Homola Linked Open Data and Its Potential for CREDO

slide-39
SLIDE 39

Scientific Data in LOD

How can science benefit from LOD?

◮ Scientific datasets are already published in LOD ◮ Specific scientific domain ontologies

Especially in :

◮ Medicine ◮ Biology ◮ Linguistics

(see example)

Our next target:

◮ Information security ◮ CREDO (?)

Martin Homola Linked Open Data and Its Potential for CREDO

slide-40
SLIDE 40

Gamification

Martin Homola Linked Open Data and Its Potential for CREDO

slide-41
SLIDE 41

Gamification

Gamification: application of game-like features in non-game context.

Martin Homola Linked Open Data and Its Potential for CREDO

slide-42
SLIDE 42

Gamification

Gamification: application of game-like features in non-game context. Gamification has been successfully applied in:

◮ User engagement ◮ Organizational productivity ◮ Learning ◮ Crowdsourcing ◮ Employee recruitment ◮ Ease of use ◮ Physical exercise ◮ Traffic violations ◮ Voter apathy

Martin Homola Linked Open Data and Its Potential for CREDO

slide-43
SLIDE 43

Gamification for engagement/learning

Problem: Students fail to engage in activities which are not compulsory

Martin Homola Linked Open Data and Its Potential for CREDO

slide-44
SLIDE 44

Gamification for engagement/learning

Problem: Students fail to engage in activities which are not compulsory

Martin Homola Linked Open Data and Its Potential for CREDO

slide-45
SLIDE 45

Thank You for Your Attention!

Questions?

Martin Homola Linked Open Data and Its Potential for CREDO